pt-kill doesn't reconnect if its connection is lost

Bug #941469 reported by Steve McTeague
12
This bug affects 2 people
Affects Status Importance Assigned to Milestone
Percona Toolkit moved to https://jira.percona.com/projects/PT
Fix Released
High
Daniel Nichter

Bug Description

Percona Toolkit v.2.0.3
Mysql: mysqld Ver 5.0.92-community-log for unknown-linux-gnu on x86_64 (MySQL Community Edition (GPL))

When pt-kill is daemonized and then viewing its log, find this error message:

Error getting SHOW PROCESSLIST: DBD::mysql::st execute failed: MySQL server has gone away [for Statement "SHOW FULL PROCESSLIST"] at /usr/bin/pt-kill line 3633.

Thank you

Related branches

Revision history for this message
Steve McTeague (smcteague) wrote :

Realize now that running two concurrent pt-kill scripts could be causing this issue.

tags: added: pt-kill
Revision history for this message
Baron Schwartz (baron-xaprb) wrote :

Steve, do you have reason to believe that one instance was killing the other?

I think the bug report could also be summarized as "pt-kill doesn't reconnect if its connection is lost." OK?

Revision history for this message
Steve McTeague (smcteague) wrote : Re: [Bug 941469] Re: pt-kill: error getting processlist; mysql server gone away

Baron, I agree with your updated bug description.

I stated this only because it stopped once I retained only one instance running.  I previously had two and was experiencing this issue.

Is this function supposed to be able to run in more than one instance?  I ask because I am not super talented in writing perl matching syntax and cannot always wrap all desired events, triggers, etc. into one elegant line of code.  Also, would other pt- functions run concurrently and nicely together?

________________________________
 From: Baron Schwartz <email address hidden>
To: <email address hidden>
Sent: Thursday, March 8, 2012 2:07 PM
Subject: [Bug 941469] Re: pt-kill: error getting processlist; mysql server gone away

Steve, do you have reason to believe that one instance was killing the
other?

I think the bug report could also be summarized as "pt-kill doesn't
reconnect if its connection is lost." OK?

--
You received this bug notification because you are subscribed to the bug
report.
https://bugs.launchpad.net/bugs/941469

Title:
  pt-kill: error getting processlist; mysql server gone away

Status in Percona Toolkit:
  New

Bug description:
  Percona Toolkit v.2.0.3
  Mysql: mysqld Ver 5.0.92-community-log for unknown-linux-gnu on x86_64 (MySQL Community Edition (GPL))

  When pt-kill is daemonized and then viewing its log, find this error
  message:

  Error getting SHOW PROCESSLIST: DBD::mysql::st execute failed: MySQL
  server has gone away [for Statement "SHOW FULL PROCESSLIST"] at
  /usr/bin/pt-kill line 3633.

  Thank you

To manage notifications about this bug go to:
https://bugs.launchpad.net/percona-toolkit/+bug/941469/+subscriptions

Revision history for this message
Steve McTeague (smcteague) wrote :

Baron, I agree with your updated bug description.

I stated this only because it stopped once I retained only one instance running. I previously had two and was experiencing this issue.

Is this function supposed to be able to run in more than one instance? I ask because I am not super talented in writing perl matching syntax and cannot always wrap all desired events, triggers, etc. into one elegant line of code. Also, would other pt- functions run concurrently and nicely together?

Revision history for this message
Baron Schwartz (baron-xaprb) wrote : Re: pt-kill: error getting processlist; mysql server gone away

The tools should coexist nicely, or at least in a well defined way. Making pt-kill reconnect if it loses its connection, especially when daemonized, is an important feature.

tags: added: error-recovery
Changed in percona-toolkit:
status: New → Triaged
summary: - pt-kill: error getting processlist; mysql server gone away
+ pt-kill doesn't reconnect if its connection is lost
tags: added: percona-23049
Changed in percona-toolkit:
milestone: none → 2.1.3
status: Triaged → Confirmed
importance: Undecided → High
Changed in percona-toolkit:
status: Confirmed → In Progress
assignee: nobody → Daniel Nichter (daniel-nichter)
Changed in percona-toolkit:
status: In Progress → Fix Committed
Changed in percona-toolkit:
status: Fix Committed → Fix Released
Revision history for this message
Shahriyar Rzayev (rzayev-sehriyar) wrote :

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PT-299

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.